\section{E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$ Class Template Reference}
\label{classmlpack_1_1gmm_1_1EMFit}\index{E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$@{E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$}}


This class contains methods which can fit a \doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM} to observations using the EM algorithm.  


\subsection*{Public Member Functions}
\begin{DoxyCompactItemize}
\item 
\textbf{ E\+M\+Fit} (const size\+\_\+t max\+Iterations=300, const double tolerance=1e-\/10, Initial\+Clustering\+Type clusterer=\+Initial\+Clustering\+Type(), Covariance\+Constraint\+Policy constraint=\+Covariance\+Constraint\+Policy())
\begin{DoxyCompactList}\small\item\em Construct the \doxyref{E\+M\+Fit}{p.}{classmlpack_1_1gmm_1_1EMFit} object, optionally passing an Initial\+Clustering\+Type object (just in case it needs to store state). \end{DoxyCompactList}\item 
const Initial\+Clustering\+Type \& \textbf{ Clusterer} () const
\begin{DoxyCompactList}\small\item\em Get the clusterer. \end{DoxyCompactList}\item 
Initial\+Clustering\+Type \& \textbf{ Clusterer} ()
\begin{DoxyCompactList}\small\item\em Modify the clusterer. \end{DoxyCompactList}\item 
const Covariance\+Constraint\+Policy \& \textbf{ Constraint} () const
\begin{DoxyCompactList}\small\item\em Get the covariance constraint policy class. \end{DoxyCompactList}\item 
Covariance\+Constraint\+Policy \& \textbf{ Constraint} ()
\begin{DoxyCompactList}\small\item\em Modify the covariance constraint policy class. \end{DoxyCompactList}\item 
void \textbf{ Estimate} (const arma\+::mat \&observations, std\+::vector$<$ Distribution $>$ \&dists, arma\+::vec \&weights, const bool use\+Initial\+Model=false)
\begin{DoxyCompactList}\small\item\em Fit the observations to a Gaussian mixture model (\doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM}) using the EM algorithm. \end{DoxyCompactList}\item 
void \textbf{ Estimate} (const arma\+::mat \&observations, const arma\+::vec \&probabilities, std\+::vector$<$ Distribution $>$ \&dists, arma\+::vec \&weights, const bool use\+Initial\+Model=false)
\begin{DoxyCompactList}\small\item\em Fit the observations to a Gaussian mixture model (\doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM}) using the EM algorithm, taking into account the probabilities of each point being from this mixture. \end{DoxyCompactList}\item 
size\+\_\+t \textbf{ Max\+Iterations} () const
\begin{DoxyCompactList}\small\item\em Get the maximum number of iterations of the EM algorithm. \end{DoxyCompactList}\item 
size\+\_\+t \& \textbf{ Max\+Iterations} ()
\begin{DoxyCompactList}\small\item\em Modify the maximum number of iterations of the EM algorithm. \end{DoxyCompactList}\item 
{\footnotesize template$<$typename Archive $>$ }\\void \textbf{ serialize} (Archive \&ar, const unsigned int version)
\begin{DoxyCompactList}\small\item\em Serialize the fitter. \end{DoxyCompactList}\item 
double \textbf{ Tolerance} () const
\begin{DoxyCompactList}\small\item\em Get the tolerance for the convergence of the EM algorithm. \end{DoxyCompactList}\item 
double \& \textbf{ Tolerance} ()
\begin{DoxyCompactList}\small\item\em Modify the tolerance for the convergence of the EM algorithm. \end{DoxyCompactList}\end{DoxyCompactItemize}


\subsection{Detailed Description}
\subsubsection*{template$<$typename Initial\+Clustering\+Type = kmeans\+::\+K\+Means$<$$>$, typename Covariance\+Constraint\+Policy = Positive\+Definite\+Constraint, typename Distribution = distribution\+::\+Gaussian\+Distribution$>$\newline
class mlpack\+::gmm\+::\+E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$}

This class contains methods which can fit a \doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM} to observations using the EM algorithm. 

It requires an initial clustering mechanism, which is by default the K\+Means algorithm. The clustering mechanism must implement the following method\+:


\begin{DoxyItemize}
\item void Cluster(const arma\+::mat\& observations, const size\+\_\+t clusters, arma\+::\+Row$<$size\+\_\+t$>$\& assignments);
\end{DoxyItemize}

This method should create \textquotesingle{}clusters\textquotesingle{} clusters, and return the assignment of each point to a cluster. 

Definition at line 45 of file em\+\_\+fit.\+hpp.



\subsection{Constructor \& Destructor Documentation}
\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a030d5debc5e91c51e0f72f64ff39f3de}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!E\+M\+Fit@{E\+M\+Fit}}
\index{E\+M\+Fit@{E\+M\+Fit}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{E\+M\+Fit()}
{\footnotesize\ttfamily \textbf{ E\+M\+Fit} (\begin{DoxyParamCaption}\item[{const size\+\_\+t}]{max\+Iterations = {\ttfamily 300},  }\item[{const double}]{tolerance = {\ttfamily 1e-\/10},  }\item[{Initial\+Clustering\+Type}]{clusterer = {\ttfamily InitialClusteringType()},  }\item[{Covariance\+Constraint\+Policy}]{constraint = {\ttfamily CovarianceConstraintPolicy()} }\end{DoxyParamCaption})}



Construct the \doxyref{E\+M\+Fit}{p.}{classmlpack_1_1gmm_1_1EMFit} object, optionally passing an Initial\+Clustering\+Type object (just in case it needs to store state). 

Setting the maximum number of iterations to 0 means that the EM algorithm will iterate until convergence (with the given tolerance).

The parameter force\+Positive controls whether or not the covariance matrices are checked for positive definiteness at each iteration. This could be a time-\/consuming task, so, if you know your data is well-\/behaved, you can set it to false and save some runtime.


\begin{DoxyParams}{Parameters}
{\em max\+Iterations} & Maximum number of iterations for EM. \\
\hline
{\em tolerance} & Log-\/likelihood tolerance required for convergence. \\
\hline
{\em clusterer} & Object which will perform the initial clustering. \\
\hline
{\em constraint} & Constraint policy of covariance. \\
\hline
\end{DoxyParams}


\subsection{Member Function Documentation}
\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a399ea7835ff966f467d13416046c7fa0}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Clusterer@{Clusterer}}
\index{Clusterer@{Clusterer}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Clusterer()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily const Initial\+Clustering\+Type\& Clusterer (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the clusterer. 



Definition at line 113 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a79abb61d6e17dced3b7bffcec5a5a83a}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Clusterer@{Clusterer}}
\index{Clusterer@{Clusterer}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Clusterer()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily Initial\+Clustering\+Type\& Clusterer (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}}



Modify the clusterer. 



Definition at line 115 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a2abb870b3266e1ce64e43ed16f2537f6}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Constraint@{Constraint}}
\index{Constraint@{Constraint}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Constraint()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily const Covariance\+Constraint\+Policy\& Constraint (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the covariance constraint policy class. 



Definition at line 118 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a05ddd1b39294c15e05c8d922b336ee2f}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Constraint@{Constraint}}
\index{Constraint@{Constraint}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Constraint()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily Covariance\+Constraint\+Policy\& Constraint (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}}



Modify the covariance constraint policy class. 



Definition at line 120 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a2a2f9c9c82f1e94ee7aad7d759193b97}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Estimate@{Estimate}}
\index{Estimate@{Estimate}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Estimate()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily void Estimate (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{observations,  }\item[{std\+::vector$<$ Distribution $>$ \&}]{dists,  }\item[{arma\+::vec \&}]{weights,  }\item[{const bool}]{use\+Initial\+Model = {\ttfamily false} }\end{DoxyParamCaption})}



Fit the observations to a Gaussian mixture model (\doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM}) using the EM algorithm. 

The size of the vectors (indicating the number of components) must already be set. Optionally, if use\+Initial\+Model is set to true, then the model given in the means, covariances, and weights parameters is used as the initial model, instead of using the Initial\+Clustering\+Type\+::\+Cluster() option.


\begin{DoxyParams}{Parameters}
{\em observations} & List of observations to train on. \\
\hline
{\em means} & Vector to store trained means in. \\
\hline
{\em covariances} & Vector to store trained covariances in. \\
\hline
{\em weights} & Vector to store a priori weights in. \\
\hline
{\em use\+Initial\+Model} & If true, the given model is used for the initial clustering. \\
\hline
\end{DoxyParams}
\mbox{\label{classmlpack_1_1gmm_1_1EMFit_ac0f0d7036d31c3f590afd3ca9d061425}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Estimate@{Estimate}}
\index{Estimate@{Estimate}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Estimate()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily void Estimate (\begin{DoxyParamCaption}\item[{const arma\+::mat \&}]{observations,  }\item[{const arma\+::vec \&}]{probabilities,  }\item[{std\+::vector$<$ Distribution $>$ \&}]{dists,  }\item[{arma\+::vec \&}]{weights,  }\item[{const bool}]{use\+Initial\+Model = {\ttfamily false} }\end{DoxyParamCaption})}



Fit the observations to a Gaussian mixture model (\doxyref{G\+MM}{p.}{classmlpack_1_1gmm_1_1GMM}) using the EM algorithm, taking into account the probabilities of each point being from this mixture. 

The size of the vectors (indicating the number of components) must already be set. Optionally, if use\+Initial\+Model is set to true, then the model given in the means, covariances, and weights parameters is used as the initial model, instead of using the Initial\+Clustering\+Type\+::\+Cluster() option.


\begin{DoxyParams}{Parameters}
{\em observations} & List of observations to train on. \\
\hline
{\em probabilities} & Probability of each point being from this model. \\
\hline
{\em means} & Vector to store trained means in. \\
\hline
{\em covariances} & Vector to store trained covariances in. \\
\hline
{\em weights} & Vector to store a priori weights in. \\
\hline
{\em use\+Initial\+Model} & If true, the given model is used for the initial clustering. \\
\hline
\end{DoxyParams}
\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a420770944a5b0c7a852c4ec372c4a2d1}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Max\+Iterations@{Max\+Iterations}}
\index{Max\+Iterations@{Max\+Iterations}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Max\+Iterations()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily size\+\_\+t Max\+Iterations (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the maximum number of iterations of the EM algorithm. 



Definition at line 123 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_acda675ab4ab86b95c92bc33bc391a61b}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Max\+Iterations@{Max\+Iterations}}
\index{Max\+Iterations@{Max\+Iterations}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Max\+Iterations()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily size\+\_\+t\& Max\+Iterations (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}}



Modify the maximum number of iterations of the EM algorithm. 



Definition at line 125 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a68e832cb064e3b7ca978d8e5911cf700}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!serialize@{serialize}}
\index{serialize@{serialize}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{serialize()}
{\footnotesize\ttfamily void serialize (\begin{DoxyParamCaption}\item[{Archive \&}]{ar,  }\item[{const unsigned int}]{version }\end{DoxyParamCaption})}



Serialize the fitter. 



Referenced by E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$\+::\+Tolerance().

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a7b5af5c1a84c507cbaa7f999ea5a4fda}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Tolerance@{Tolerance}}
\index{Tolerance@{Tolerance}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [1/2]}}
{\footnotesize\ttfamily double Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption}) const\hspace{0.3cm}{\ttfamily [inline]}}



Get the tolerance for the convergence of the EM algorithm. 



Definition at line 128 of file em\+\_\+fit.\+hpp.

\mbox{\label{classmlpack_1_1gmm_1_1EMFit_a3d9fac84af16250f5a3689692e8f2173}} 
\index{mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}!Tolerance@{Tolerance}}
\index{Tolerance@{Tolerance}!mlpack\+::gmm\+::\+E\+M\+Fit@{mlpack\+::gmm\+::\+E\+M\+Fit}}
\subsubsection{Tolerance()\hspace{0.1cm}{\footnotesize\ttfamily [2/2]}}
{\footnotesize\ttfamily double\& Tolerance (\begin{DoxyParamCaption}{ }\end{DoxyParamCaption})\hspace{0.3cm}{\ttfamily [inline]}}



Modify the tolerance for the convergence of the EM algorithm. 



Definition at line 130 of file em\+\_\+fit.\+hpp.



References E\+M\+Fit$<$ Initial\+Clustering\+Type, Covariance\+Constraint\+Policy, Distribution $>$\+::serialize().



The documentation for this class was generated from the following file\+:\begin{DoxyCompactItemize}
\item 
/var/www/mlpack.\+ratml.\+org/mlpack.\+org/\+\_\+src/mlpack-\/3.\+3.\+1/src/mlpack/methods/gmm/\textbf{ em\+\_\+fit.\+hpp}\end{DoxyCompactItemize}